Soon after we published the functional analysis of the first NaV1.7 mutations from patients with erythromelalgia, I was deluged with an avalanche of emails, letters, and telephone calls from around the world. They were from people with chronic pain. Some of these people had erythromelalgia, and some of them had mutations of NaV1.7 that we had not seen before. Here were new clues. But there also were entreaties, requests for help. Especially touching were the enquiries from parents of children in pain. In the beginning, I felt nearly helpless. I was navigating a large, complex sea.
We were looking for rare experiments of nature in which the gene for NaV1.7 had gone awry, with the hope that each new genetic mistake would teach us something new. To do this, we established a network of physicians and scientists not only in North America, but also in Europe and Asia, who helped us to sift out, from an overall population of more than two billion, the most instructive cases. Investigation of each new patient took a long time, and we needed to prioritize so that we could focus on the ones we could learn from. The work-up for each mutation required months of work by teams of scientists that moved like clockwork within my laboratory. Sulayman Dib-Hajj led a team of molecular biologists that carefully re-created, in the laboratory, DNA for each mutation. Then our physiologists went to work. The 2006 paper by Han et al. (Han et al. 2006), for example, described a sporadic de novo mutation that we encountered in a child with normal parents. Undoubtedly, many mutations in NaV1.7 arose in this way rather than being inherited from parents.
As researchers working in a competitive arena, my team was used to working at a “let’s be first” pace. We weren’t quite frenzied but, until we had completed our analysis of the first four or five mutations, we moved forward in “hurry-up” mode. Each mutation taught us something new. In 2006 our analysis of the S241T mutation taught us that the size of the substituted residue was important in determining pathogenicity of the mutation (Lampert et al. 2006). Soon after that, molecular modeling gave us a picture of the ring of four amino acids, at the cytoplasmic mouth of the channel pore, that act as a gate, stabilizing the channel in the closed state (Lampert et al. 2008). Mutations described in our papers by Cheng et al. (2008) and Han et al. (2009) taught us that the magnitude of the change in channel function caused by the mutation plays a role in determining the age of onset of pain—mutations that produced large hyperpolarizations of activation caused pain beginning in infancy, while mutations that produced smaller shifts in activation caused pain with a later onset. We found two mutations that not only cause inherited erythromelalgia, but also sensitize the NaV1.7 channel to particular drugs (Choi et al. 2009; Fischer et al. 2009), a finding that pointed us toward development of personalized, genomically guided pharmacotherapy for pain (Yang et al. 2012). In 2011 we discovered a mutation which causes erythromelalgia not by substituting an incorrect amino acid for the correct one, but by deleting an amino acid (Cheng et al. 2011). We also learned that there are polymorphisms, or minor misweaves, in the gene for NaV1.7 that, while not causing disease within themselves, impose increased risk for developing neuropathic pain after nerve injury (Estacion et al. 2009). Each gene variant showed us something about the disease or the channel. And, as we looked at the entire group of mutations, there was another, broader observation: In all of these cases, hyperactive NaV1.7 channels caused pain but did not cause epileptic seizures, an observation that reinforced our notion of a major functional role of NaV1.7 in peripheral nerves, but not in the brain.
Following our description of NaV1.7 gain-of-function mutations in inherited erythromelalgia, a research group at University College London described another set of gain-of-function mutations of NaV1.7 that cause a second, distinct clinical disorder, paroxysmal extreme pain disorder (PEPD) (Fertleman et al. 2006). PEPD, which had previously been called familial rectal pain disorder (Fertleman et al. 2007), presents with a clinical picture very different than that from inherited erythromelalgia. These mutations were unique in that they tended to cause disease, not by enhancing activation, but by impairing the process of channel inactivation which prevents channels from operating for a brief period after stimulation. Impaired inactivation makes more channels available for operation. In patients with PEPD, lower body stimulation, particularly stimulation close to the rectum, triggers excruciating rectal pain which, later in life, migrates to the area around the eyes and jaw. The reason for this peculiar pattern of pain is not known, but, importantly, patients with PEPD usually respond well to treatment with the sodium channel blocker carbamazepine, making the diagnosis clinically important.
We also learned that, while inherited erythromelalgia and PEPD each present striking and very different clinical features, they are part of a continuum. In 2008, we described a patient with an “overlap” syndrome and clinical features of both disorders (Estacion et al. 2008). The NaV1.7 mutation in this patient had multiple effects on the function of the channel, including the enhanced activation characteristic of erythromelalgia mutations and the impaired inactivation usually seen with PEPD mutations. We subsequently described several additional mutations which caused atypical forms of erythromelalgia.
Inherited erythromelalgia and PEPD represented graphic examples of “gain-of-function mutations,” with heightened function of the mutation resulting in excruciating pain. In 2006, patients with “loss-of-function” mutations of NaV1.7 and “channelopathy-associated insensitivity to pain” were found. These patients failed to make functional NaV1.7 channels. The affected individuals felt no pain. The first patient described was a teenager, from tribal Pakistan, who helped to support his family with “street performances” in which he would injure himself by putting blades through his limbs or walking on hot coals, feeling no pain. His family carried a mutation that impaired their ability to produce NaV1.7 channels. This family was replete with individuals who had experienced painless fractures, painless burns, painless tooth extractions, and painless childbirth (Cox et al. 2006). Other, similar families were soon identified, each family carrying a “null” mutation so that they did not produce functional NaV1.7 channels (Ahmad et al. 2007; Goldberg et al. 2007). What was remarkable about these families was not that there was a diminished sense of pain, but rather that they did not feel any pain.
These “human knock-out” mutations reminded us that, in the absence of pain, people do not learn to protect themselves by limiting their activities in the way most of us do. Individuals lacking NaV1.7, with channelopathy-associated insensitivity to pain, will continue to play a game of soccer, for example, after sustaining a fracture, and accumulate multiple unhealed injuries. NaV1.7 was known to be present in olfactory sensory neurons which are essential for our sense of smell, and, consistent with this, anosmia—loss of the sense of smell—was also observed in these loss-of-function patients. Aside from this, as would be predicted for a channel that does not play a major role within the brain, people with channelopathy-associated insensitivity to pain showed no other signs of brain dysfunction.
By 2006, just two years after the first NaV1.7 mutations were found, we had been able to move from gene, to altered channel protein, to pain-signaling DRG neurons that scream when they should be silent. We knew that gain-of-function mutations of NaV1.7 produce hyperexcitability of DRG neurons that causes excruciating pain. And we knew by 2006 that loss-of-function of NaV1.7 produces inability to sense pain. We were in the unusual situation of understanding NaV1.7 through both gain-of-function and loss-of-function.